The Computational Biology Support Component (Leader, Thomas Kepler) will be organized into two teams: one team at Duke under Thomas Kepler, and the other at Los Alamos led by Bette Korber and Alan Perelson. Independently and jointly, they will work to surmount the two roadblocks: HlV-1 diversity and the inability to induce broad neutralizing antibodies (BnAbs) to HIV-1 envelope. Both sites will be involved in and collaborate on statistical analyses and immunogen design, but each will have its own primary focus. The Duke team will focus on the analysis and modeling of affinity maturation toward protective antibody phenotypes, including the inference of clonal histories back to the unmutated common ancestor and integrated analysis of structural, phenotypic, and genetic antibody data. The Los Alamos team will focus on polyvalent and mosaic immunogen design to overcome HIV diversity and assist in modeling affinity maturation and antibody repertoire. Together, the teams will statistically evaluate all results obtained from testing of the various designs as well as perform co-evolutionary analyses of BnAbs and transmitted/founder viral sequences together with the Viral Biology SRSC. Both teams will contribute to immunogen design and the development of different approaches. Specific Aims Aim 1. Statistical analysis of HIV broadly neutralizing antibodies (BnAb) and vaccine-induced antibody and virus evolution. Aim 2. Computational design of vaccine antigens for the induction and maturation of protective antibody responses.